FW: MAF Upgrad

[Frederic Breitwieser]

>How about running two MAF sensors in parallel?
>I'm sure you'd need some kind of happy box to interface the two
>and send out an integrated signal.  If there were enough interest,
>this could be a source of income for somebody ...

I've been thinking about doing this with my twin-turbo system I'm building
- each OEM MAF could flow enough air to support 3 cyl's (I have a 4.1L
six), however putting the outputs in parallel was not as easy as I had
anticipated.  I expected the outputs to be a straight voltage representing
the flow, much like how the other sensors work - the GM units I have at
least (from a 1986 3.8L LeSabre) vary their frequency output, at a constant
voltage, to represent the flow.  So summing would be more than a simple
operational amplifier addition type of thing.

I built a simple circuit using three LM9400's and one TL084 Op Amp, however
was not pleased with the results, it was beyond unstable.  But I will
describe in case you or someone else is enthused enough to come up with a
better solution :)

Basically, each MAF sensor is attached to the LM9400, configured as a
Freq-->Voltage converter.  Those two outputs then connect to the TL084
(high impedence op-amp) and are summed, then using a second portion of the
TL084 I scaled the output down a little bit, then ran it into another
LM9400, which was configured as a voltage to frequency converter, then to a
third portion of the TL084 to clean and stabalize the signal.  The output
was okay, except there are certain inherent design flaws.

I one of the MAFs go bad, and send no signals, the output of the circuit
does lower, but doesn't stop entirely.  Both MAFs would have to fail in
order for the ECM to have a clue (a GM ECM at least) that something is
seriously wrong.  Also, by the same token, averaging the outputs is
dangerous, what happens if one turbo and wastegate goes nuts, and pumps 30
PSI into the intake, and the other side sits at 3psi?  The imbalance could
be very bad for the engine, as one side would detonate severely.

Anyway, how I tested this was to put two MAFs physically in series using
duct tape, attached an intake hose with duct take, and connect it to the
outlet of my shop vac, also with duct tape.  I then attached the A/C cord
of the shop vac to a light dimmer, and slowed and sped up the shop vac
accordingly (until the dimmer overheated).

Anyway, I have abandoned the idea of summing...  someone here brightly
suggested creating a toggle switch circuit (using a CD4066 maybe?) to
switch between the two Mafs, thus alternating the readings every 1/4 of a
second, or something more or less along those lines.

More than likely, I'm going to have to find a giant MAF and run both
turbo's through it, using Y-Pipes.  Though I must admit I'm very tempted to
dismantle one of them and make a bigger housing, and repeat the shop-vac
test and see what readings I get (of course, with a better dimmer <Grin>).

Hope I didn't bore everyone!


Frederic Breitwieser
Bridgeport, CT 06606
http://www.xephic.dynip.com/

1993 Supercharged Lincoln Continental
1989 4-Door Softtop Humvee (Hummer)
2000 Buick GTP (Mid-engine track car)

---